Profile avatar
wandedob.bsky.social
Associate professor (biträdande professor) in neuroimaging. Research in data analysis, image processing, statistics, deep learning, federated learning, synthetic images. Tweets in Swedish and English. Libertarian.
268 posts 1,186 followers 1,334 following
Prolific Poster
Conversation Starter

Applied for 2 million EUR from the Swedish Cancer society, for early detection of lung cancer using deep learning. We will use the huge SCAPIS dataset (30 000 subjects), and train AI models through federated learning between Linköping, Lund, Umeå and Stockholm. Fingers crossed.

Submitting a grant application from a train, because I like living on the edge

Alarming new study in JAMA modeling measles in the US: If we stay at CURRENT vaccination rates, measles will become endemic again over roughly 20 years. If rates drop by even 10 percent (and they're sliding), there would be 11 MILLION cases over the next 25 years. jamanetwork.com/journals/jam...

Strange experience when you contact the university travel agency and they reply that there is no longer an agreement with the university, and the procurement officers at the university say that there is an agreement

Wonder how many grants and papers that have been written on airports

10 days left to the deadline and I have not decided if I should write 0, 1 or 2 applications

I just want to tip my hat to the crack team of White House economists who were able to discover--in just a few short days--that the U.S. is dependent on China for smartphones, computers and semiconductors.

If a funding agency has a call for "generative AI agents for cancer" what do you think they want?

Our federated training of medical segmentation models between Linköping, Lund and Umeå is now up and running again. Waiting for more cities to join.

13 years ago I nailed my PhD thesis

Today we write in the magazine Curie about the lack of feedback on grant applications, and that it becomes weird when funding agencies require open science but don't provide a single word of feedback on applications. www.tidningencurie.se/debatt/oppen...

Mycket bra förslag Lägg 100 miljarder kronor på att bygga ut Sveriges grundforskning i en historisk expansion utan motstycke. Rekrytera flera hundra framstående amerikanska akademiker som alltmer förtvivlat söker nya möjligheter att bedriva sin forskning www.tidningencurie.se/debatt/forsk...

I would fund this research

Hmm

Any journals that accept papers about brain imaging AND philosophy?

4 grant rejections so far in 2025, and not a single word of feedback

Is there any paper showing that matching patient group and control group by age and sex is not sufficient, but that you have to include covariates for age and sex in regressions to fully remove such effects?

Has anyone seen this before? Paper accepted without any revision.

Right: vision in grant application Left: final results in paper

Our debate article about lack of feedback on grant applications has now been rejected 4 times, without any feedback.

So when do you start paying your reviewers?

Found reviewer 2 of our last paper

Yes our federation between Linköping, Lund and Umeå is up again, training a segmentation network with local data in each city.

Cool visit at Sweden's first nuclear reactor

I was just asked to prove that I know english for an american visa, guess that writing 75 papers, living in the US for 2 years, and teaching in english for 15 years is not enough.

Join us at the 6th Annual Biomedical Engineering Conference! We are pleased to invite you to BME@LiU, where we highlight research in the field of biomedical engineering. This conference is an excellent opportunity for researchers, companies, organisations, students, and clinicians to connect,

Claude managed to give me new ideas for my grant application, cool

Bedrövligt att en så liten del av Vetenskapsrådets forskningsfinansiering går till forskning inom teknik. Varför satsar man inte lika mycket på teknik, som på medicin och naturvetenskap?

Is there a word for when a grant application is rejected and you are sort of happy that you dont need to start a new project?

What if you paid the reviewers instead, just a thought www.nature.com/articles/d41...

Have you ever encountered the problem of self plagiarism when publishing several papers? How did you solve it or avoid it?

PI who says "We will publish all papers as open access in Nature" vs lab manager who knows the cost of open access in Nature journals

RELICT: A Replica Detection Framework for Medical Image Generation Good that the memorization problem finally is getting recognition arxiv.org/abs/2502.17360

Svenska forskningsfinansiärer förespråkar i allt större utsträckning öppen vetenskap, men granskning av ansökningar om forskningsmedel sker fortfarande bakom stängda dörrar utan återkoppling till de sökande.

Svenska forskningsfinansiärer förespråkar i allt större utsträckning öppen vetenskap, men granskning av ansökningar om forskningsmedel sker fortfarande bakom stängda dörrar utan återkoppling till de sökande.

I don't understand funding agencies who score grant applications, use the scores to decide who will get a grant, and then refuse to send out the scores to the applicants. What are you afraid of?

New dataset on AIDA datahub, by Christian Jamtheim Gustafsson and colleagues in Lund. A radiotherapy medical image dataset called 🌟 LUND-PROBE 🌟, featuring 467 unique prostate cancer MRI and synthetic CT volumes, radiotherapy dose distributions, and delineated organs! 🎉

Dear prospective grad students: when you use ChatGPT to write your email inquiring about our program, and it hallucinates faculty members for you to work with who don't exist,,, it doesn't look great.